The purpose of this competing renewal is to document the effectiveness of short-term interventions for non-psychiatrically disturbed firesetters and to test existing approaches used in clinical and community settings. The content of intervention is based on findings from the previous assessment and follow-up grant study of firesetting and nonfiresetting children (MH39976). One hundred and twenty child firesetters (aged 7-12) will be randomly assigned to one of three conditions: 1) Fire Safety Education (FSE), 2) Psychosocial Treatment intervention will be short-term, executed by trained specialists for each approach (firefighters, therapists) using program manuals, monitored to ensure therapeutic integrity, and evaluated using multiple measures from multiple sources. Pre-post comparisons will be supplemented with both 3- and 12-month follow-up data. The primary aim is to compare the impact of each intervention on the child involvement with fire and the severity of firesetting, and secondarily, on primary psychosocial correlates of firesetting (e.g., fire risk factors, interpersonal repertoire and antisocial behavior, and parent/family stress). It is hypothesized that FSE and PT, relative to FOT, will be superior in reducing fire involvement at follow-up both at home and in the community, and that PT will be superior to FSE or FOT in reducing the psychosocial correlates of firesetting. The second aim is to derive the predictors of firesetting recidivism for each condition and moderating variables to document client x intervention interactions using background and posttreatment variables (e.g., fire safety knowledge, curiosity about fire, antisocial behavior). As such, it is believed that this proposal represents the first such treatment outcome evaluation study for this population and may facilitate the matching of clients to treatment.